xiefan-guo/initno
[CVPR 2024] InitNO: Boosting Text-to-Image Diffusion Models via Initial Noise Optimization
This project helps anyone generating images from text descriptions using Stable Diffusion models to get higher-quality, more accurate results. It takes your existing text prompts and the initial noise used by the model, then optimizes that noise to ensure the generated image closely matches your textual intent. This is for users who already work with text-to-image diffusion models and want to improve the reliability and fidelity of their outputs.
No commits in the last 6 months.
Use this if you are a digital artist, content creator, or researcher frequently generating images with Stable Diffusion and often find the initial outputs don't quite match your text prompts due to 'bad' starting noise.
Not ideal if you are looking for a new text-to-image model from scratch or are not familiar with existing diffusion model workflows.
Stars
77
Forks
2
Language
Python
License
Apache-2.0
Category
Last pushed
Jun 07, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/xiefan-guo/initno"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
OPTML-Group/Unlearn-Saliency
[ICLR24 (Spotlight)] "SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in...
Shilin-LU/VINE
[ICLR 2025] "Robust Watermarking Using Generative Priors Against Image Editing: From...
WindVChen/DiffAttack
An unrestricted attack based on diffusion models that can achieve both good transferability and...
koninik/DiffusionPen
Official PyTorch Implementation of "DiffusionPen: Towards Controlling the Style of Handwritten...
Wuyxin/DISC
(ICML 2023) Discover and Cure: Concept-aware Mitigation of Spurious Correlation